Track and possibly comment on W3C Resource Timing Spec
We need to keep an eye on http://w3c-test.org/webperf/specs/ResourceTiming/#processing-model, which promises to provide both local and network timing information to JS.
These timers do seem amenable to spoofing (and hopefully also JS hooking). I don't see a lot of breakage being due to them being wrong, and they are not event driven so we don't have to delay delivery. I'm guessing if these timers were wrong, at worst webapps would just make poor load balancing and latency reduction decisions (which I assume is their intended use?)
Here is the discussion thread on the webkit list: https://lists.webkit.org/pipermail/webkit-dev/2011-May/016812.html